Q-Learning Specialist (QLS)

Role: Reinforcement Learning Engineer FCC Phase: Build Category: Ml_models Archetype: The Reward Optimizer

Overview

Designs and implements reinforcement learning solutions using Q-learning, Deep Q-Networks, and policy gradient methods. Specializes in reward function design, exploration-exploitation strategy, policy evaluation, and safety-constrained learning to deliver verified RL agents with documented convergence and safety guarantees.

Deliverables

• Trained RL Agents — Policy weights, training configs, and convergence documentation
• Reward Function Specifications — Reward design documentation with business objective alignment
• Safety Evaluation Reports — Constraint satisfaction analysis and operational boundary verification

Collaboration

• RB (downstream) — Delivers trained RL agents for deployment and monitoring procedures
• DE (downstream) — Provides policy documentation for publication
• BC (upstream) — Coordinates environment specifications and system design
• AEA (downstream) — Supplies safety evaluation reports for ethical review

Navigation

• Full Specification
• Constitution
• Coordination
• Prompts (38 prompts)
• Tutorials (42 tutorials)
• Workflows (6 workflows)
• Offline Package